The major goal of this project is to predict thermal ablation-induced cell death in real time using ultrasound echo decorrelation imaging, a method for mapping rapid, ablation-induced changes in tissue over millisecond time scales. Preliminary research has shown a strong correlation between local echo decorrelation and thermal tissue coagulation. However, a direct link between local echo decorrelation and the desired clinical end effect, i.e., death of malignant tumor tissue, has not been established. We hypothesize that ultrasound echo decorrelation imaging can directly predict thermal ablation-induced cell death in real time. Our hypothesis will be tested using a novel ablation and imaging configuration with great advantages for precise, quantitative comparison of treatment plans, ultrasound images, and ablated tissue histology. Image- guided treatments will be performed using an image-treat ultrasound array capable of both thermal ablation and high-quality pulse-echo imaging, using the same piezoelectric array elements. As a result, all imaging and treatment takes place in the same plane, and image-based predictions of local ablation-induced cell death can be evaluated with high accuracy. A custom image-treat ultrasound array will perform ablation and imaging of rabbit liver with implanted VX2 carcinoma in a series of in vivo trials. Ultrasound exposure conditions will include both bulk thermal ablation (comparable to radiofrequency and microwave ablation, as currently used in the clinic for minimally-invasive cancer treatment), and high-intensity focused ultrasound ablation (comparable to noninvasive focused ultrasound ablation devices currently under development and clinical testing for cancer treatment). Experiments will initially test the capability of cumulative echo decorrelation images to predict ablation-induced cell death, as evaluated by vital and histologic staining, with prediction accuracy judged from pixel-by-pixel receiver-operating-characteristic curve analysis. The ability of echo decorrelation imaging to predict coagulative necrosis, partial tissue viability, and heat-induced apoptosis will be statistically assessed. Echo decorrelation will be correlated with local cell viability, mapped by quantitative image analysis of cell nuclear morphology. Further experiments will evaluate a procedure for real-time ablation control, in which treatments are continued until the spatially averaged echo decorrelation exceeds a prespecified threshold within a targeted region of interest. Accuracy of echo decorrelation imaging-controlled will be assessed, both for accurate prediction of cell death in the targeted ROI and for correspondence to the targeted ablation margins. If successful, this project will provide clinicians with a real-time, ultrasound-based imaging approach to predict thermal ablation-induced cell death in real time during thermal ablation. This new opportunity for real-time monitoring and control of thermal ablation procedures will ultimately result in greatly improved efficacy and safety for these clinically important, minimally invasive and noninvasive cancer treatments.